I'm trying to load the model using load_model but it's giving me the following error in google colab.

code

from keras.models import load_model
model = load_model('keras/model.h5')

error

ValueError                                Traceback (most recent call last)
<ipython-input-48-a10a91d5d77d> in <module>()
      1 from keras.models import load_model
----> 2 model = load_model('keras/model.h5')

/usr/local/lib/python3.6/dist-packages/keras/engine/saving.py in load_model(filepath, custom_objects, compile)
    417     f = h5dict(filepath, 'r')
    418     try:
--> 419         model = _deserialize_model(f, custom_objects, compile)
    420     finally:
    421         if opened_new_file:

/usr/local/lib/python3.6/dist-packages/keras/engine/saving.py in _deserialize_model(f, custom_objects, compile)
    219         return obj
    220 
--> 221     model_config = f['model_config']
    222     if model_config is None:
    223         raise ValueError('No model found in config.')

/usr/local/lib/python3.6/dist-packages/keras/utils/io_utils.py in __getitem__(self, attr)
    300             else:
    301                 if self.read_only:
--> 302                     raise ValueError('Cannot create group in read only mode.')
    303                 val = H5Dict(self.data.create_group(attr))
    304         return val

ValueError: Cannot create group in read only mode.

Can someone please help me in understanding what is happening? How do I load this model?

Traceback (most recent call last):
  File "./demo_image.py", line 247, in <module>
    model = get_testing_model()
  File "/home/xy/share/reidPrj/keras_Realtime_Multi-Person_Pose_Estimation/model.py", line 184, in get_testing_model
    stage0_out = vgg_block(img_normalized, None)
  File "/home/xy/share/reidPrj/keras_Realtime_Multi-Person_Pose_Estimation/model.py", line 29, in vgg_block
    x = conv(x, 64, 3, "conv1_1", (weight_decay, 0))
  File "/home/xy/share/reidPrj/keras_Realtime_Multi-Person_Pose_Estimation/model.py", line 20, in conv
    bias_initializer=constant(0.0))(x)
  File "/home/wuyong/anaconda3/envs/tensorflow/lib/python3.5/site-packages/keras/engine/topology.py", line 528, in __call__
    self.build(input_shapes[0])
  File "/home/wuyong/anaconda3/envs/tensorflow/lib/python3.5/site-packages/keras/layers/convolutional.py", line 125, in build
    raise ValueError('The channel dimension of the inputs '
ValueError: The channel dimension of the inputs should be defined. Found `None`.

Seem like something wrong with img_input_shape. So I chance 177 line of model.py to this:

img_input_shape = (None, None, 3) ==>
   img_input_shape = (674, 712, 3)

Then new error occur:

Traceback (most recent call last):
  File "./demo_image.py", line 247, in <module>
    model = get_testing_model()
  File "/home/xy/share/reidPrj/keras_Realtime_Multi-Person_Pose_Estimation/model.py", line 184, in get_testing_model
    stage0_out = vgg_block(img_normalized, None)
  File "/home/xy/share/reidPrj/keras_Realtime_Multi-Person_Pose_Estimation/model.py", line 40, in vgg_block
    x = pooling(x, 2, 2, "pool2_1")
  File "/home/xy/share/reidPrj/keras_Realtime_Multi-Person_Pose_Estimation/model.py", line 24, in pooling
    x = MaxPooling2D((ks, ks), strides=(st, st), name=name)(x)
  raise ValueError(err.message)
ValueError: Negative dimension size caused by subtracting 2 from 1 for 'pool2_1/MaxPool' (op: 'MaxPool') with input shapes: [?,356,1,128].

What's wrong with that? What shape img_input_shape should be ?

I've finished python version of rmpe_server. It is 4.5 times faster than C++ version(140 pic/s vs 30 pics/sec on my server) code is approx. 2 times shorter and it is python, not c++

I made pull request not to immediately join it to main repo, at this point I want somebody except me looks on my code, test it, try to train model, etc.

update: during training loss is becoming NaN, need to find a bug update2: fixed

Hi @michalfaber, thanks for the fantastic work. I was trying to see if I can still train with local data only, without augmentation. Apparently it says that "use_client_gen = False" is deprecated in the code, and a brief try shows that the format in locally generated h5 file is not compatible with what specified in DataIterator. So my question is that, was there once a working version of for training with local data that you can upload, or maybe you can shed some light on how I should modify the code if I want to train with local data without augmentation? Thanks!

I tried to estimate the pose of the subjects in an image. After running the program with arguments. It seems like it is running for forever. It never gives output. I tried it on i5 6th gen CPU.

https://github.com/michalfaber/keras_Realtime_Multi-Person_Pose_Estimation/blob/ff482e118ace9c9e6e3d0291fb4f2cca385d0ae7/training/dataset.py#L129

Hi,

Great repo! But is this correct? you're applying the mask to the input image, when the paper says to only apply it to the PAF and heatmap outputs during training.

I couldn't find a similar snippet in https://github.com/ZheC/Realtime_Multi-Person_Pose_Estimation

If this is intentional, why is it needed and what effect did it have?

When I try to run the inspect_dataset.ipynb file with the hdf5 file val_generated_dataset.h5, which I create with the generate_hdf5.py script. I ran into this error. KeyError: "Unable to open object (Object 'label' doesn't exist)"

In generate_hdf5.py, I see there is only create one group "dataum".
grp = h5.create_group("datum")

So seems the group "label" not been created when creating the val_generated_dataset.h5. if the inspect_dataset.ipynb file not been updated after the code change of generate_hdf5.py ?

I downloaded the zip directly and run the "demo_image.py", which returned:

Traceback (most recent call last): File "D:/Python/keras_Realtime_Multi-Person_Pose_Estimation-master/keras_Realtime_Multi-Person_Pose_Estimation-master/demo_image.py", line 7, in <module> from config_reader import config_reader File "D:\Python\keras_Realtime_Multi-Person_Pose_Estimation-master\keras_Realtime_Multi-Person_Pose_Estimation-master\config_reader.py", line 1, in <module> from configobj import ConfigObj ModuleNotFoundError: No module named 'configobj'

I followed the demo.ipynb notebook to obtain the key points. I am trying to label each key point detected in the image. But, I am not able to figure out how to do that from the data in all_peaks.

The all_peaks list is something like,

[[(417, 162, 0.23725834488868713, 0)], [(438, 264, 0.22009609639644623, 1)], [(360, 260, 0.22483864426612854, 2)], [(310, 435, 0.22168569266796112, 3)], [(302, 596, 0.23836150765419006, 4)], [(511, 270, 0.2024109661579132, 5)], [], [], [(388, 580, 0.18829642236232758, 6)], [(381, 824, 0.2216421663761139, 7)], [], [(503, 577, 0.19702039659023285, 8)], [(528, 828, 0.21746164560317993, 9)], [], [(406, 140, 0.2475488781929016, 10)], [(444, 141, 0.23833531141281128, 11)], [], [(487, 159, 0.22596435248851776, 12)]]

From what I have understood, the last element in each item of the list is the label index for the keypoint. But when a certain feature is not detected, the item is empty. So I am not able to figure out how to label the keypoints.

Can anyone help me out with this?

I ran this project on windows platform. I sync coco data follow instruction, but when I executed generate_marks.py. It's crash. How to fix that. Please suggest me.

While testing my py_rmpe_server, I noticed rmpe_dataset_server is never randomly resizes images. I.e. main person on image always fixed size.

params.scale_prob=1;

  float dice = Rand(RAND_MAX) / static_cast <float> (RAND_MAX);
  float scale_multiplier;

  //actually with scale_prob==1 condition is always true -- anatolix
  if(dice > param_.scale_prob) {
    img_temp = img_src.clone();
    scale_multiplier = 1;
  }
  else {
    float dice2 = Rand(RAND_MAX) / static_cast <float> (RAND_MAX);
    scale_multiplier = (param_.scale_max - param_.scale_min) * dice2 + param_.scale_min; //linear shear into [scale_min, scale_max]
  }

I see that you successfully converted the classic COCO model from the openpose repo, but have never seen the Body_25 model? Have you had any luck with this in particular with converting the PReLU layers as that seems to be the point that would cause issues....

Thanks for your effort. I have some question about batch_dataflow function.

1. There is 'x'. what is the meaning of the x?
2. There are [0, 1, 2], [3, 4, 3, 4, 3, 4...]. I wonder why are you grouping [0, 1, 2] and [ 3, 4...] respectively.
3. Why [3, 4, 3, 4 ... ] is repeatedly??

Hi all, I have surveyed this repo. There are many relu layer for activation, but I donot find activation layer for output.

The output heatmap and pafmap should be in 0~1. If there is no activation layer for output, how can we ensure the values scaled into this range?

loading annotations into memory... Done (t=0.00s) creating index... index created! loading annotations into memory... Done (t=0.00s) creating index... index created! Loading dataset /content/keras_Realtime_Multi-Person_Pose_Estimation/dataset/train ... Loading image annot 0/12 Loading dataset /content/keras_Realtime_Multi-Person_Pose_Estimation/dataset/val ... Loading image annot 0/2 Traceback (most recent call last): File "train_pose.py", line 208, in batch_df = batch_dataflow(df, batch_size) File "/content/keras_Realtime_Multi-Person_Pose_Estimation/training/dataset.py", line 212, in batch_dataflow df = BatchData(df, batch_size, use_list=True) File "/usr/local/lib/python3.6/dist-packages/tensorpack/dataflow/common.py", line 95, in init assert batch